Drugs play a dual role in our understanding of neuropsychiatric disorders. First, they are used clinically to alleviate symptoms and (ideally) to reverse aberrant pathophysiology. Second, they provide a molecular "challenge" to the CNS that responds and provides valuable information about the ways in which relevant phenotypes are modulated at the molecular level. Thus, they provide an excellent opportunity to link high level phenotypes and low-level molecular and cellular networks. Unfortunately, the relationships between drugs, genes, and phenotypes are disperse and not amenable to computational analysis. Integrating multiple data sources relevant to neuropsychiatric drugs, their associated indications and side effects, and their molecular targets in the context of genetic interactions will be a powerful way to both generate new hypotheses about how these drugs act and which genetic circuits they affect. Thus, the primary goal of this component project Is to develop and apply informatics methods for using drug-associated data to inform models of neuropsychiatric disease risk, occurrence and treatment.